167 个 · 工具 · AI 核心
桌面端 AI 伴侣，集成 Live2D 形象、自主编码 Agent（AgentLoop）、语义记忆与确定性主动引擎——真实进程沙箱（bwrap）、权限门控工具执行、local-first（SQLite + ONNX）。Desktop AI companion with Live2D avatar, autonomous coding agent (AgentLoop), semantic memory, and a deterministic proactivity engine — real process sandboxing (bwrap), permission-gated tool execution, local-first (SQLite + ONNX).
为 AI 编程 agent 构建 Claude Code prompt 模板，包含系统 prompt、工具 prompt、记忆管理与多 agent 协调。Build Claude Code prompt templates for AI coding agents, with system prompts, tool prompts, memory, and multi-agent coordination
MCP 桥接，让 Claude Code 和其他宿主将实现工作委托给 OpenAI Codex CLI。MCP bridge that lets Claude Code and other hosts delegate implementation to the OpenAI Codex CLI
Codex Autonomous Objective Loops 2026：强制审查与验证的 AICodex Autonomous Objective Loops 2026: AI with Mandatory Review & Verification
真正好用的本地文档 → Markdown 转换器 | Local-first AI-ready document converter: PDF/Word/PPT/Excel/EPUB/HTML → clean Markdown
The Convergence Gap 的代码与 artifact：指令微调模型何时收敛到下一 token 预测Code and artifacts for The Convergence Gap: when instruction-tuned models settle on next-token predictions.
一道 Python 守门机制：在不分类提示词的前提下，阻止不可信的 agent 数据去构造敏感的工具调用参数。A Python gate that prevents untrusted agent data from authoring sensitive tool-call arguments—without classifying prompts.
🔍 使用多种 prompt 技巧在多步数学问题上分析 Mistral-7B 模型的数学推理能力。🔍 Analyze the mathematical reasoning abilities of the Mistral-7B model using diverse prompting techniques on multi-step math problems.
浏览器本地的 HTTP 请求转换器，支持 cURL、Fetch、Python、Go、Rust 和 Node.js。Browser-local HTTP request converter for cURL, Fetch, Python, Go, Rust, and Node.js.
一个面向 Codex custom subagent 的纪律严明、跨平台的编排系统。A disciplined, cross-platform orchestration system for Codex custom subagents
2026 年由 AI 驱动的 Things 3 CLI 自动化工具——macOS 智能任务管理器AI-Powered Things 3 CLI Automation Tool 2026 – Smart Task Manager for macOS
面向 AI 辅助软件工程的循证门控式系统设计，以研究为依据。Research-informed, evidence-gated system design for AI-assisted software engineering.
在小型 LLM 中植入认知美德的激活引导实验——三项发现与一个失败模式数据集（DOI 见 README）Activation-steering experiments on installing epistemic virtues in small LLMs — three findings + a failure-mode dataset (DOIs in README).
面向空间 LLM 与地理空间 AI Agent 的生产工程模式——验证优先架构、地理空间 RAG、prompt-to-spatial-SQL、tool routing 与评估。Production engineering patterns for spatial LLMs and geospatial AI agents — validation-first architecture, geospatial RAG, prompt-to-spatial-SQL, tool routing, and evaluation.
macOS 菜单栏 App：统计各 AI Agent 工具的 token 用量与花费，显示订阅额度，并用状态灯实时反映任务运行状态。纯本地只读。
生产级 LLM 网关：兼容 OpenAI 的 API，支持语义缓存，可在 Claude/OpenAI/Ollama 之间进行智能路由与成本分析。Production LLM gateway: OpenAI-compatible API, semantic caching, intelligent routing across Claude/OpenAI/Ollama, cost analytics.
使用 TurboQuant 和 MLX 在 Windows 上运行 Qwen3.5 语言模型，实现快速的本地推理。Run the Qwen3.5 language model on Windows using TurboQuant and MLX for fast local performance.
AI 驱动的 CORBA/IDL 接口自动化 —— 从自然语言规约到实时 ORB 绑定，无需手写 stub；MIT ORB core 基于开放 OMG wire 规范构建。AI-driven CORBA/IDL interface automation — from natural-language spec to a live ORB binding, with no hand-written stubs. MIT ORB core built on the open OMG wire specification.
MeterSphere v2.10 个人开发分支 — 一站式开源持续测试平台。工作流引擎(Flowable 7) · 需求池(从0到1) · 微前端(qiankun→micro-app) · AI知识库 · 测试跟踪，含个人笔记与创意项目。
借助这款轻量、原生的 Text Services Framework 输入法编辑器在 Windows 上输入孟加拉语，无需后台进程。Type Bangla on Windows with this lightweight, native Text Services Framework Input Method Editor that operates without background processes.
Agentic-SDLC 审计工具，整个工具完全通过其所记录的流水线自身构建：规划、agent 构建、对抗式 AI 审查、CI 卡点、人工授权合并。针对 AI agent 开发规范输出确定性评分报告。Agentic-SDLC audit tool built entirely through the pipeline it documents: plan, agent build, adversarial AI review, CI gate, human-authorized merge. Deterministic scored reports on AI-agent development discipline.
一个窗口管理你拥有的每个编码 Agent。Claude Code、Codex、Gemini CLI、Qwen、Ollama 等各占独立终端面板。支持 Windows 和 macOS，免费，无需账号，无需服务器。One window for every coding agent you own. Claude Code, Codex, Gemini CLI, Qwen, Ollama and more, each in its own terminal pane. Windows and macOS, free, no accounts, no server.
升级 Grand Theft Auto V（GTA V）画面表现，增强真实光照、天气效果与改进贴图，带来终极 4K 视觉体验。Upgrade Grand Theft Auto V graphics with realistic lighting, weather effects, and improved textures for a definitive 4K visual experience.
在双 RTX 3090 GPU 上以 262K 上下文长度并发流运行 Qwen3.6-27B，开发工作已迁移至 club-3090。Run Qwen3.6-27B at 262K context with concurrent streams on dual RTX 3090 GPUs. Development moved to club-3090.
成本感知的 GitHub Copilot 工作流，用于清晰、有界且经过独立评审的软件变更。Cost-aware GitHub Copilot workflows for clarified, bounded and independently reviewed software changes.
针对 PX4 飞控软件的覆盖率引导事务语义审计研究 artifactResearch artifact for coverage-guided transactional semantic auditing of PX4 flight-control software.
使用支持 macOS 和 Windows 平台的开源工具高效构建与定制机器学习模型Build and customize machine learning models efficiently with an open-source tool that supports macOS and Windows platforms.
开源、本地优先的足球数据、分析与内容工作流。Open-source, local-first football data, analysis, and content workflow.
通过单一严格的 YAML 清单统一管理本地 LLM 运行时——支持状态查看、健康检查、启动、停止，基于原生 systemd 与 Docker。Manage local LLM runtimes from one strict YAML manifest — status, doctor, start, stop over native systemd and Docker.
💳 使用 XGBoost 检测金融交易欺诈，结合针对不平衡数据集与复杂特征的高级优化技术💳 Detect fraudulent financial transactions using XGBoost. Optimize performance with advanced techniques for imbalanced datasets and complex features.
硕士论文（KCL）：在攻击者-机器留出划分下重新衡量 IoT 入侵检测，并测试基于 LLM 的报文预测是否能提升性能。移除泄露后，macro-F1 从 0.90 降至 0.60。MSc dissertation (KCL): re-measuring IoT intrusion detection under an attacker-machine holdout, and testing whether LLM-based packet prediction improves it. macro-F1 0.90 -> 0.60 once leakage is removed.
🎨 利用 GPT、Gemini 等模型，AI 驱动的学术图表一键生成与定制工具🎨 Generate academic diagrams effortlessly with this AI-driven tool, leveraging models like GPT and Gemini for seamless creation and customization.